Using image error level analyzers to analyze gemstone photos

[Mico]

RTfrog brough up in a different thread regarding photography and the use of an image error level analyzer to help evaluate pictures of gemstones - as we all know they can be very hard to photograph.

I'd like more information on this, which would be ideal if we maybe used tweaked PSers photos, and non-vendor photos, to help diminish any backlash or tempers that may flare during the discussion.

This is a purely educational topic without any malice or vendor bias, please keep it a happy place . I'd very much like to hear from both consumers and vendors as it was mentioned by some vendors in the other thread that sometimes minimal tweaks are necessary to reflect the stone's true character.

 

[Alnitak]

Re: Using image error level analyzers to analyze gemstone ph

I'm afraid there really aren't any analyzers one can use to do this sort of thing.

I know this thread was started by the use of a tool in another thread, but the reality is that user did not know what the tool did or how to understand the results. Here's a quote directly from that website about what the *can* do:

"ELA measures the amount of change during a JPEG resave. When a digital photo is edited, the modified portions will have a different error level potential compared to the rest of the picture. Splices, drawing, and significant edits are usually visible as a significantly different error level potential.

There is a difference between real and authentic. A real photo of a forged document or a staged situation will not appear unusual under ELA. This is because the picture is real, even if the subject of the photo is not authentic. ELA does not identify the authenticity or other attributes related to the picture's subject.

ELA also does not detect all forms of digital manipulation; it only identifies differences in the JPEG compression rate. Digital modifications that do not significantly alter the error level potential, such as a minor color adjustment over the entire picture, may not be detected by ELA."

That pretty much means its useless for the purpose of detecting image changes that matter to a colored stone buyer. Why? Because the photographer can shoot in a raw format--native data as it comes out of the camera--and then make all of the changes they want to before saving as a jpeg. Because its the first time the image is saved, there is *nothing* for the algorithm to detect, since there are no differences in jpeg compression between two or more saves of a file as a jpeg-compressed image. Even then, the group that puts the software out there makes it clear that things like color adjustments to a whole image may not be detected.

The reality is that no matter how good a photo is, no matter how much we try--there are too many variables at play to know that your eyes will perceive a gem exactly as its seen in a photo online. What you need is a vendor return policy that is reasonable, so you can examine the stone and return it if its not what you expected/wanted based on the photos.

Jeff

 

[mastercutgems]

Re: Using image error level analyzers to analyze gemstone ph

Excellent words of wisdom Alnitak;

As I have run into all these problems years ago and finally just take all my photos outside in subdued lighting. I know it is not representative of fluorescent or other lighting conditions; but I cut for a living and do not have the time to set up all the fancy lighting conditions.

But I tell everyone how I take pictures; show it on a video on my front page, etc.... But at the end of the day I have found some of the most important things are to be kind, honest, and give that excellent return policy with no re-stocking fees, etc... That is the best I can do as I have learned over 30 years ago I will not make it through this life pleasing everyone...

I know many on this forum do not like my way of photography but sadly it is the best I can do for now

Thank you for putting light to the subject of photography of gems as it really is a daunting task to represent a gem for its true color value especially on shade shifting gems and not give it a glamor shot that will truly disappoint the buyer when they open the paper.

Thank you and Most respectfully;

Dana Reynolds
ASG
Certified Supreme Master gem Cutter
#96CGE42

 

[bcavitt]

Re: Using image error level analyzers to analyze gemstone ph

Mico -

Our resident expert in gem/diamond photography is kenny. I'm hope he'll chime in on this subject.

 

[endless_summer]

Re: Using image error level analyzers to analyze gemstone ph

Excellent words of wisdom Alnitak;

As I have run into all these problems years ago and finally just take all my photos outside in subdued lighting. I know it is not representative of fluorescent or other lighting conditions; but I cut for a living and do not have the time to set up all the fancy lighting conditions.

But I tell everyone how I take pictures; show it on a video on my front page, etc.... But at the end of the day I have found some of the most important things are to be kind, honest, and give that excellent return policy with no re-stocking fees, etc... That is the best I can do as I have learned over 30 years ago I will not make it through this life pleasing everyone...

I know many on this forum do not like my way of photography but sadly it is the best I can do for now

Thank you for putting light to the subject of photography of gems as it really is a daunting task to represent a gem for its true color value especially on shade shifting gems and not give it a glamor shot that will truly disappoint the buyer when they open the paper.

Thank you and Most respectfully;

Dana Reynolds
ASG
Certified Supreme Master gem Cutter
#96CGE42

Dana,

I appreciate your candor & the way that you take pictures. As a consumer without a wealth of experience, I find it's easiest for me to read color from a couple simple cell phone pictures in outdoor & indoor lighting. I think that the glamour shots of some vendors are really very beautiful, especially when they are equally as representative of the stone (I can't imagine how much work goes into that), but simple works too and makes me feel like I have a better sense of what the stone will look like in hand.

 

[Alnitak]

Re: Using image error level analyzers to analyze gemstone ph

Dana,

I appreciate your candor & the way that you take pictures. As a consumer without a wealth of experience, I find it's easiest for me to read color from a couple simple cell phone pictures in outdoor & indoor lighting. I think that the glamour shots of some vendors are really very beautiful, especially when they are equally as representative of the stone (I can't imagine how much work goes into that), but simple works too and makes me feel like I have a better sense of what the stone will look like in hand.

Sometimes I'll mix it up and show photos of both--studio shots and shots taken outdoors on the back of my hand under indirect light.

Jeff

 

[FrekeChild]

Re: Using image error level analyzers to analyze gemstone ph

But I tell everyone how I take pictures; show it on a video on my front page, etc.... But at the end of the day I have found some of the most important things are to be kind, honest, and give that excellent return policy with no re-stocking fees, etc... That is the best I can do as I have learned over 30 years ago I will not make it through this life pleasing everyone...

I know many on this forum do not like my way of photography but sadly it is the best I can do for now

Thank you for putting light to the subject of photography of gems as it really is a daunting task to represent a gem for its true color value especially on shade shifting gems and not give it a glamor shot that will truly disappoint the buyer when they open the paper.

Thank you and Most respectfully;

Dana Reynolds
ASG
Certified Supreme Master gem Cutter
#96CGE42

Dana, I just wanted to say thank you. I've never actually bought anything from you, but you've always been very kind in your correspondence with me.

And I wanted to say that I appreciate this point of view - I would rather buy a stone that looks ok in a photo and looks beautiful in real life than buy into a glamour shot and be disappointed when I get the package.

 

[endless_summer]

Re: Using image error level analyzers to analyze gemstone ph

Dana,

I appreciate your candor & the way that you take pictures. As a consumer without a wealth of experience, I find it's easiest for me to read color from a couple simple cell phone pictures in outdoor & indoor lighting. I think that the glamour shots of some vendors are really very beautiful, especially when they are equally as representative of the stone (I can't imagine how much work goes into that), but simple works too and makes me feel like I have a better sense of what the stone will look like in hand.

Sometimes I'll mix it up and show photos of both--studio shots and shots taken outdoors on the back of my hand under indirect light.

Jeff

Jeff - that's probably the best of both worlds, and hand shots are always, always helpful!

 

[endless_summer]

Re: Using image error level analyzers to analyze gemstone ph

But I tell everyone how I take pictures; show it on a video on my front page, etc.... But at the end of the day I have found some of the most important things are to be kind, honest, and give that excellent return policy with no re-stocking fees, etc... That is the best I can do as I have learned over 30 years ago I will not make it through this life pleasing everyone...

I know many on this forum do not like my way of photography but sadly it is the best I can do for now

Thank you for putting light to the subject of photography of gems as it really is a daunting task to represent a gem for its true color value especially on shade shifting gems and not give it a glamor shot that will truly disappoint the buyer when they open the paper.

Thank you and Most respectfully;

Dana Reynolds
ASG
Certified Supreme Master gem Cutter
#96CGE42

Dana, I just wanted to say thank you. I've never actually bought anything from you, but you've always been very kind in your correspondence with me.

And I wanted to say that I appreciate this point of view - I would rather buy a stone that looks ok in a photo and looks beautiful in real life than buy into a glamour shot and be disappointed when I get the package.

+1!

 

[Alnitak]

Re: Using image error level analyzers to analyze gemstone ph

The real challenge is to provide a "glamor shot" that doesn't oversell the stone. That's the best of all worlds, but very hard to achieve and impossible with some gems or some cuts of some gems. My use of multiple images for some stones is my "cop-out" from spending any more time trying to get that elusive shot.

That said...I always want my gems to be even better in the hand. I think that even with a glamour shot this is possible, as a photo cannot capture all of the brilliance and play of color of a finely faceted stone. I've started posting HD videos of some of my stones to try and capture that when I can, but that's also very time-consuming.

Jeff

 

[Mico]

I actually think videos are an awesome idea. I wonder if some kind of lighting box can be made that the gem can move through under different lighting situations while filming...

Meh, one can dream

 

[RTFrog]

Re: Using image error level analyzers to analyze gemstone ph

Hi Mico,

There is a lot more to Image Error Level Analyzers, different algorithms, and other techniques to detail image manipulation or edits. Each time you want to evaluate an image or suite of images, it needs to be done on a case-by-case basis. There are some great resources about how and why this works, including ways it can be circumvented here:

http://www.wired.com/images_blogs/threatlevel/files/bh-usa-07-krawetz.pdf

Each different analysis can help provide tools in determining how a picture should or could look.

Some pictures are truly worth a thousand words.

Please also look into Histograms and if there is any other data present within the image - as this can reveal more details in the EXIF file formats.

It would be extremely useful to have a standardized color setup to capture a stones truly representative color in various lighting conditions. This would be ideal as long as users did not then post-process and manipulate hues/saturations and other tones.

I’ll be glad to continue this discussion more after the threats and menacing emails have calmed down.

 

[kenny]

Re: Using image error level analyzers to analyze gemstone ph

RTfrog thanks for introducing me to ELA.
I had no clue about it.

In other situations ELA, while not perfect, is clearly a useful tool.
Here it serves no purpose because photoshopping is not only okay it's essential as I will explain.
Worse, many here still think that ALL Photoshopping as bad and this ELA tool being used to attempt to 'prove' image modification feeds into and fans the flames of that ignorance.
Photoshop is just a tool and is as inherently innocent as a phone, a knife, or a car - three other tools that can be used for good or evil.

Let me explain a bit about myself, and why say Photoshopping is always okay.
I'm a geeky gem photographer.
I had a Nikon D200 body, then a D7000.
I now use a Nikon D600 and my new D800 will arrive Thursday.
To get a 3 mm gem to fill up the frame in the camera so I use all of my sensor's pixels, I use a 30-year old Nikon PB-4 Bellows that can hold the lens up to 10" from the body.
I use a 30 year old Nikkor 105mm f2.8 micro lens because it has the manual aperture ring required for use with the bellows.
The old bellows does not have the electrical contacts needed for today's bodies to talk to today's lenses to tell them what aperture to go to.
When I need even more magnification I'll reverse a 24mm at the end of the bellows and I can blow up a girdle inscription like it was under a microscope.



My brand new Nikkor 105mm f2.8G with all those nice features, sharper glass and Nano coating arrived today.
No I can't use it on the bellows but it will be nice for pics of larger things like a ring.
I use a light tent and have produced some images that are complimented here by civilians and pros.

So, I may not be a pro but I know my way around a fine macro set up.

I always post-process.
Always.
Every pic!

I use post processing to bring the camera's output back to the truth … to the best of my ability given the limitations of my various monitors etc.
A fine camera, macro lens, the best lights and a competent photographer is not enough to get perfectly-exposed pics out of a camera when the subject is so quirky.
All cameras are dumb and they lie.
They can't help it because they are expected to do what's impossible.
They are programed to set every exposure so a print of the pic will reflect 18% of the light that strikes it.

If you take one pic of a white horse in mid-day full-sun standing in a field of snow and another pic of a black horse standing in front of a black wall after sun down both pics from the best cameras in the world will come out looking a similar gray.
That's stupid!
What comes directly out of camera is … well … lies … unless the brightness of the important parts of your subject also happen to average something near that 18% reflectance.

Humans are smarter than cameras.
Humans can brighten up the incorrectly-grayish pic of the white horse and darken down the incorrectly-grayish pic of the black horse.
Photoshopping is the right thing to do.
Then both pics cab tekk the truth.

Same with gem pics.
Backgrounds and gems are all over the tonal map from dark to light. (Let's not even get started on hue.)
I can shoot a white diamond on a white (or a dark) background or a dark garnet on a dark (or a light) background and the camera will make all of these pics look of similar medium brightness, 18%.
Cameras don't know some subjects are supposed to come out bright and others are supposed to be dark so they expose all of them for 18% average reflectance.
How could they know any better, really?

Unfortunately many people are poorly informed about photography and the need for post-processing.
The problem stated way back in 1900 when the original Brownie camera made photography seem easy for the masses.

Even back then retouching film or paper prints could be as innocent as filling in dust spots or as evil as removing a politician from a pic to change history.
Unfortunately the evil use is what sticks in everyone's mind when it comes to then retouching and today's Photoshopping …

Since a naughty vendor Photoshpping colored stones on an Internet site is so harmful to us buying gems the broad-brush negative reaction to all Photoshopping is staggering here on PS.

OH NO KENNY PHOTOSHOPS HIS PICS!
LET'S CRUCIFY HIM!
DOESN'T HE KNOW THAT WHATEVER COMES DIRECTLY OUT OF A CAMERA IS SACRED, PURE AND TRUE!?!


Nuff said.

 

[RTFrog]

Re: Using image error level analyzers to analyze gemstone ph

Hi Kenny,

IELA and other techniques come out over the years that often are not discussed or picked up by a huge audience because they are highly technical, subject to upset some or have controversial features, and generally do not reveal things too far out of the ordinary for those that can compose masterful realistic photos. I have found your work with gemstone to be stunning and inspirational in trying to create something anywhere close.

Now about photoshop and image manipulation. It's just like everything else - some people are good or bad at it.

I can see where there are instances of good and bad photoshop, along with necessity to adjust the gemstone to match what your eye sees. True image and color reproduction is a talent. If we didn't have a hard time with matching color, especially when applicable digitally, then there wouldn't be billions of colors digitally acceptable or possible today. The evolution of 8 to 48 bit color is absolutely stunning, especially as compared to the limited amount of time.

This thread is here to talk about IELA, or any other technique that someone uses/finds to observe digital photographic edits, and occurred from a specific instance where some images were evaluated with the algorithm along with other tools that revealed that the images were edited/manipulated from a photo editor. Even cropping an image creates technical image manipulation. Even with gemstone photography, IELA can reveal manipulation - especially if the image was adequately composed. I'm going to try and explain with images below that I have taken along with the below results through the IELA. Some were photoshopped, some were not. As a standard practice, I have stripped away EXIF data, which should help focus only on the results of IELA. The before and after shots certainly help to identify which pictures have been edited, however, when you've used the algorithm and other techniques long enough - it's just like taking pictures. The same can be said when presented by a series of pictures shot by the same camera, lens, and can account for lighting variation over a period of time. It's much easier to get a general feeling of which images have more manipulation then others.













PS - Kenny this thread isn't about crucifying photographers that photoshop. I have seen many photos composed and created by great photographers that routinely use photoshop to help enhance their photos "ethically". The problem is that others, without a great setup or equipment, rely far too often on software to adjust the images to try and create the same end result. The problem with these poor or naughty photographers is that their abuse of software can create questionable results and images that are sometimes impossible to create with their hardware setup. This thread should help provide a way to distinguish between them.

 

[kenny]

Re: Using image error level analyzers to analyze gemstone ph

RTfrog, thanks for the compliment.

Even when 'manipulation' is identified - so what?
What does it matter if the operator did stuff before or after the pic was taken?

Sure, it's nice to save time 'manipulating' things in the camera rather than in post but all that matters is the result.
My camera has settings.
Photoshop has settings.
There may be technical reasons why it is preferable to do stuff in the camera, but that does not make doing so morally superior to adjusting the same stuff in post.

My camera has exposure compensation and also a setting to make pics more 'Vivid', and has many white balance settings and even lets me do a live white balance that perfectly matches the light source used instead of assuming the bulb has color X.
Using these would cut down the time I spend Photoshopping but I don't use all of them.
Just lazy I guess … but if a PS jury is going to use IELA to put every pic on trial then perhaps I should do my 'manipulation' in the camera.

I know of a vendor who lets light enter the pavilion for their pics, which can conceal poor cut like leakage and windowing.
Some people like this style of photography.
It's pretty; it's dazzling.
Also pretty is the extremely soft focus of another vendor, which is achieved by shallow depth of field from a wide-open aperture.
Having only 3 eyelashes in focus made Hollywood starlets look glamorous and may sell diamonds to the type of customers who are pushovers for an emotional reaction but these two techniques strike me as truly manipulative.

IELA won't reveal this manipulation.
What IELA can reveal in gem photography is not a crime … but the PS Photoshop lynch mob doesn't understand this.
I'm sure I'm speaking words that vendors wish they could speak but, since they understand the dynamics of the crowd here, dare not.

Again IELA is cool and interesting but there is no legit use for it here on Pricescope.
It will only fan fires that arise from lack of education about honest photography being a two-step process … pre and post.

IELA can't reveal intent.
It can't reveal whether the knife was used for murder or for cutting a steak … only that the knife was used.

 

[RTFrog]

Re: Using image error level analyzers to analyze gemstone ph

Kenny: I believe the discussion and use of any technique to provide a consumer a more full picture of what they are looking at belongs on pricescope. It is especially important when dealing with unscrupulous vendors, and to not be silenced or threatened for providing a critique of a gemstone or image.

Here is the same gemstone:





To me the editing appears purposeful and shows the intent of image editor. Much of this may be able to be gleaned from merely looking at the image, but this type of evaluation helps provide a technical explanation for why something may look edited.

 

[kenny]

Re: Using image error level analyzers to analyze gemstone ph

Well, I tried.

 

[RTFrog]

Re: Using image error level analyzers to analyze gemstone ph

And I appreciate it, especially without becoming

I have always wanted to find ways to help understand or read photos better, especially now with gemstones. This has not been a flash in the pan, been pretty consistent about it, and something that occurs everywhere. This is one of those tools - that when used correctly can help provide additional analysis and information about photos.

I hope that the community here has other recommendations or tools/tricks/interpretations of how they read gemstones. The only tool we have on a consistent basis are consumer pictures of vendor's gemstones or photo comparisons. The pantone colors used also can help show someone what they are getting. Seeing threads calling a Spinel stoplight red - then comparing it to a stoplight was pretty awesome, especially since the camera caught both of them in the same frame.

Kenny, you would be one of the primary voices that can help guide this conversation. Again your photos are awesome and accurate. Is there any way that you would recommend an easy way to detect if a color is accurate or if image manipulation has occurred to make something unnatural?

 

[kenny]

Re: Using image error level analyzers to analyze gemstone ph

One big reason the situation is nearly hopeless is we are not photographing opaue objects.
We are photographing little boxes of windows and mirrors.
Even in real life a white diamond varies tremendously in appearance, and that's without throwing color into this can of worms.

I imagine Buddhist monks meditate on the paradox of capturing an 'accurate' pic of something that's constantly changing appearance with environment and light source.
You can't really take a pic of the mirror, only of its reflection.

Feedback from buyers on the accuracy of vendor pics is the best we have.
Even then, if customers tell us the color does not match it may be explained by a poor monitor they used to view the vendor pics.
Customers taking and posting pics to prove color mismatch is just as fraught with problems as any pics from any equipment or poster.

We have to get away from this misconception that a photograph posted on the Internet and viewed by a zillion different monitors proves anything.
There are too many things that cause pics to not be true.

But yes in extreme cases, such as the recent thread, the customer's pics were useful to demonstrate the color mismatch that even the vendor admitted to.
Monitors can be off, but not likely THAT off.

A little Photoshopping is okay.
A lot of Photoshopping is equally okay.
Heavy Photoshopping is equally okay.
All that maters is results that are as accurate as possible.
Photoshopping can move images away from or towards truth, but IELA can't show which is happening.

I'm sorry but that's why IELA is useless here on PS.
Sure, IELA can spot a smoking gun, but not whether the gun was used to commit a crime, or to stop one.

You asked for my opinion.

 

[endless_summer]

Re: Using image error level analyzers to analyze gemstone ph

The real challenge is to provide a "glamor shot" that doesn't oversell the stone. That's the best of all worlds, but very hard to achieve and impossible with some gems or some cuts of some gems. My use of multiple images for some stones is my "cop-out" from spending any more time trying to get that elusive shot.

That said...I always want my gems to be even better in the hand. I think that even with a glamour shot this is possible, as a photo cannot capture all of the brilliance and play of color of a finely faceted stone. I've started posting HD videos of some of my stones to try and capture that when I can, but that's also very time-consuming.

Jeff

It's a fair & good 'cop out' - I think it'd be very difficult to capture a stone in one picture. Videos are awesome, but I can see how that would be very time-consuming as well. Though, there is a balance to be struck because you guys can't spend all of your time taking pictures - at some point, folks just need to see the stone in person

 

[iLander]

Re: Using image error level analyzers to analyze gemstone ph

Well, we could go on and on about this (oh, wait! we already did! ) but I have to say Photoshop can be your best friend, especially with something as wiley as colored stones.

I will usually use it to crop a photo, but occasionally I will use it adjust a color to show what I see in RL. (For those of you with some color theory, I will adjust hue, but not value. If it's not bright, I won't make it brighter. Some stones require color adjusting! For example:

I have a sapphire that the AGL certified as a brown pinkish orange. Do you have any IDEA how hard that is to photograph? (Yes, I will post a thread on it soon). It shoots red. Absolutely ruby red. At NO point do I see red in real life. (the capital letters are my frustration at shooting this ridiculous stone!! ) If I were a dealer and tried to sell it with unaltered pics, the buyer would be really angry with me. Plus it's a color changer! In various lights, I see orange, hot magenta, deep pink, or even light copper with green flashes. Sometimes it closes up and it looks like an oldish penny.

Location also matters with CS. I have taken my stones to various latitudes, and they have different colors to them. The wavelengths in my yard are different than the ones in your yard. I once bought a stone from Barry, and he said it shifted to purple. I could NEVER get that sucker to show purple (another sapphire), but I did see some green. Barry never saw green. Fascinating.

And don't get me started on the limitations of digital RGB versus continuous tone emulsion, monitor calibration, tungsten versus fluorescent photo bulbs, etc. Don't forget how colors associate; if I put a red apple on a red blanket, it might look brownish. I put the apple on a white table it looks red again. It goes on for days.

Bottom line; we've all returned stones on CS, because there's no substitute for the real thing, in your hand, in your house. I think vendors do their best, and we're expecting a lot hoping they'll be dead accurate to our experience.

 

[lelser]

Re: Using image error level analyzers to analyze gemstone ph

We have a method for our photographs that makes us happy, and so far has made (most all) our clients happy.

Tom photographs the gem freehand, in mixed studio light. He doesn't use a tripod but takes often dozens of freehand shots. He selects the best one or two, crops them, and does basic dust removal and adjusts for the colour on his monitor versus the gem in his hand.

The gems come back to me, with the photos. I do the same exercise of colour on my monitor versus my hand in daylight (window and OTT lamp.)

The end result is a photo that we both feel represents the gem well, but under-represents it both is sparkle and colour. The photo needs to be good enough to attract a buyer, but when s/he opens the box I want the response to be "wow! That's nicer than the photo." So far, that's the bulk of the feedback I've gotten from the PS'ers who've bought from me. Yes, we're photoshopping the colour, but only to correct the camera's colour biases.

It's tricky and it takes time. My "glamour shots" are done by a professional photographer in Seattle, but I'd never use those to sell.

Cheers,

Lisa
www.lisaelser.com

 

[digdeep]

Re: Using image error level analyzers to analyze gemstone ph

I have one of those sapphires and I LOVE it even though the seller described it as brown/purple because the photo showed those colors........when he pulled it (much to his surprise) it was yellow, then green, then a few other shades. He called it a mood stone because it shifted so much! When I took a photo it was a more saturated yellow with purple reflections. On some days the silk makes it sleepy and on other days the silk disappears and I have a beautiful droplet in front of me. I didn't think silky sleepiness could disappear, yet it does!! Bottom line, NOTHING replaces having a stone in your hands and looking at it! Sometimes looking at it again, and again, and again........ it's a wonderful relationship isn't it???

 

[PrecisionGem]

Re: Using image error level analyzers to analyze gemstone ph

Im not sold on this Image Error analyzer. Here is a photo that was taken RAW as a Nikon NEF file, converted to a JPG and reduced in size cropped, and of course my water mark added to it. Nothing else, no tweaks of any kind. Based on the analyzer image, you would think I "photoshopped the hell" out it.

I don't own Photoshop by the way, I use a program called Aperture on the Mac.

 

[Alnitak]

Re: Using image error level analyzers to analyze gemstone ph

Im not sold on this Image Error analyzer. Here is a photo that was taken RAW as a Nikon NEF file, converted to a JPG and reduced in size cropped, and of course my water mark added to it. Nothing else, no tweaks of any kind. Based on the analyzer image, you would think I "photoshopped the hell" out it.

I don't own Photoshop by the way, I use a program called Aperture on the Mac.

Gene, if you read the description of the tool and what it can detect, its not really useful in the way its been represented here on the PS forums. I've done the same real-world tests as you and had the same results.

Jeff

 

[RTFrog]

Re: Using image error level analyzers to analyze gemstone ph

Kenny: Thank you for your opinion. There's a reason that IELA exists - along with practical applications in some instances. There has been extensive research done on the topic, but it still would be nice to have an automatic hue/saturation plugin or device to help convey color regardless of lighting and monitor differences.

This has been some pretty constructive and informative discussion so far. I do not believe that there is as great of an issue when vendor's go to such great care and pride in their representations of their work.

iLander: I agree that most vendors do their best. This thread reflects the care that goes into overall photo composition.

Lisa: That's a pretty extensive process! I really like how you don't use the glamour shots to sell along with under-representing the stone. This makes complete sense and does add to that wow factor. I've experienced it many times with another vendor from here, and every time it is a joy to receive them.

Gene: Nah, your image looks real for many reasons. I'm not going to modify your image and post it here, trying to avoid more hoopla. But if you go into aperture and edit the highlights, shadows, hue, and saturation... such as to make it mainly purple with a blue undertone on that gorgeous tanzanite, save a copy of it and run it through that analysis again, you'll see some pretty dramatic changes. The alterations should be pretty dynamic.

In general, can Vendor's recognize it when others take liberties with their photos? I'm not asking for names, just a general discussion. If not, a tool would be helpful for vendors too. Kenny discussed the lighting techniques that can occur to hide/place less emphasis on certain issues. I've been fairly able to spot when inclusions are being somewhat de-emphasized in photos. With the common industry practice of color adjustments to fairly match tones/saturation, the vendors that are not as experienced seem to be at a disadvantage.

 

[kenny]

Re: Using image error level analyzers to analyze gemstone ph

I don't own Photoshop by the way, I use a program called Aperture on the Mac.

FWIW every pic I've posted was processed with the lowly iPhoto program that comes free on every Mac.

I bought Lightroom and Elements (a stripped down cheapo version of Photoshop) but haven't installed them yet.

I second shooting in RAW if your camera can.
In post when you try to correct overexposed or underexposed parts in a JPEG file there very little data there to recover
With RAW overexposed areas that look solid white and blown out can contain a surprising amount of wonderful detail.
Same with some dark areas that look solid black.

RAW rocks!
Yes it takes up much more memory, but it's worth it.
If memory cost is an issue … shoot in RAW, do your post editing, save the modified files as a beautiful JPEGs and delete the original RAW files.

I don't convert, or SAVE AS, to JPEG.
I do a screen capture which creates a PNG file.
This gives me a more accurate pic on Pricescope.
When I used to convert to JPEG those picks looked washed out when uploaded to Pricescope.
For whatever reason the PNG files look correct on PS, though of course Pricescope's software dramatically lowers the sharpness/resolution.

 

[Karl_K]

Re: Using image error level analyzers to analyze gemstone ph

These programs find crappy jpg creation engines more than anything else.
You can do the the same thing in 20 different programs and a few will look like a total mess and maybe one or 2 pristine and the rest in between.
The programs that use the Microsoft native codec to do the conversion are going to be the worst.
Guess what? that is a lot of them because it is easy.

 

[Karl_K]

Re: Using image error level analyzers to analyze gemstone ph

When I used to convert to JPEG those picks looked washed out when uploaded to Pricescope.
For whatever reason the PNG files look correct on PS, though of course Pricescope's software dramatically lowers the sharpness/resolution.

You didn't save it in the right color space.
The screen shot does the conversion, jpg and png you have to set it to change to srgb in many programs.
Do your editing in srgb and it will work even better but you lose some color that is in the raw file.
You will not see the color anyway if you don't have a wide gamut calibrated monitor.

 

[PrecisionGem]

Re: Using image error level analyzers to analyze gemstone ph

Kenny, have a look at Pixlemator. It's a fraction of the cost of Photoshop. Aperture to me is head and shoulders above iPhoto. For some reason iPhoto confuses me, and aperture makes sense. Very nice for organizing photos.